tailor and latest version of bzr

[James Westby]

On (13/04/07 10:37), Gibson, Scott wrote:
> 
>        if (count != 1) or conflicts:
> 
> If I'm not totally wacked out in my estimation it would appear that the
> trouble occurs in tailor due to a change in bzr's branch.py file.  
> 
> "def pull" (line 1490) used to return:
> 
>             return new_count - old_count
> 
> which would provide the numeric value expected by tailor.  However the
> expected return value has since been replaced with the output from
> "class PullResult(_Result):" (line 2114).  Now the return values are
> something like:
> 
> <bzrlib.branch.PullResult object at 0xb760172c>
> 
> I made a temporary fix by just assigning count = 1 prior to error
> checking.  It has allowed me to continue migrating from bzr to svn but
> it is far from being a good solution.
> 
> Any thoughts on how to best validate the pull?  Or is it possible with
> the current pull output?

You can currently call int(count) to get the old return value. This will
work for both the old and new versions, so should be safe. However this
may not be around in a future version of bzr, and then you should access
the attribute of the PullResult that has the number of revisions.

Thanks,

James

-- 
  James Westby   --    GPG Key ID: B577FE13    --     http://jameswestby.net/
  seccure key - (3+)k7|M*edCX/.A:n*N!>|&7U.L#9E)Tu)T0>AM - secp256r1/nistp256